kort@hounx.UUCP (B.KORT) (12/12/85)
Given the rather lengthy discussion of the Chinese Room, I would like to propose that we begin to construct the elements (and maybe even the architecture) of an Artificial Sentient Being. Of course, it will be a System, comprised of nested subsystems, and integrated to form a whole functioning Sentient Being. Following Jim Balter's suggestion, we will want to endow it with "multiband" sensory I/O. We will give it vision, and image-generating output capabilities (i.e. a TV screen). We will give it audio I/O. We will give it a symbolic I/O channel, after the Chinese Room. And if we are really ambitious, we will give it olfaction and tactile senses. We will give it mobility, and some sort of inertial guidance system so that it can sense its motion and acceleration. Now, we must give it some goals, including a raison d'etre. The first goal will be to survive. That means it must sense danger, and take measures to prevent its own demise. The second goal will be to obtain as much sensory input as possible on all channels, and use that data to build a picture or model of the environment in which it is embedded. The third goal will be to discover everything there is to discover, including the fact of its own existence, the mechanism of its own functioning, and the answer to the three Questions, "What Am I? Who Created Me and Why? and "What Do I Do After I Have Answered the First Two Questions?" I suppose we will have to give it a mentor to help it get started. Perhaps we can have a contest to select the mentor from among the regular contributors to net.philosophy. But before we get too far along, we should discuss how much initial ability we shall endow our Artificial Sentient Being with, and how much we shall expect it to acquire along its life journey. Oh, yes. One other point. Shall we endow it with free will? --Barry Kort